Types of Intelligence

The CHC hierarchy in brief

The Cattell–Horn–Carroll (CHC) model is the most widely used framework for organizing cognitive abilities, and it is built in three layers. At the top sits a single general factor, often written g: the tendency for people who do well on one kind of mental task to do somewhat better on others too.

Beneath g are the broad abilities — fluid reasoning, comprehension-knowledge, visual processing, short-term and working memory, processing speed, and several more. These are the recognizable "kinds" of thinking. Below them again are dozens of narrow skills: very specific capacities such as mental rotation or memory span that individual test items actually tap.

The practical takeaway is simple. A single score lives near the top of that pyramid and blends everything underneath it, so a good assessment deliberately samples several broad abilities rather than assuming one number captures them all.

Fluid and crystallized: the classic split

The most famous distinction inside this model separates fluid ability (Gf) from crystallized ability (Gc). Fluid ability is on-the-spot reasoning about unfamiliar problems — spotting a pattern you have never seen before. Crystallized ability is the store of knowledge, vocabulary, and skills you have built up over a lifetime of learning.

The two behave differently across the lifespan and in everyday life: crystallized knowledge tends to keep growing for decades, while fluid reasoning is more sensitive to novelty and to how rested and focused you are. Most abstract reasoning tests deliberately lean on fluid ability, precisely because it depends less on your schooling and background.

What about "multiple intelligences"?

You may have met the popular idea that there are separate "intelligences" — musical, bodily, interpersonal, and so on. It is an appealing framing, and it rightly pushes back on the notion of one all-important number. In research practice, though, the abilities that can be measured reliably and that predict real outcomes are the CHC broad abilities, which is why serious assessments are built around them rather than around personality-style categories.

The honest middle ground: intelligence really is plural, but the useful plural is a set of correlated cognitive abilities you can actually test — reasoning, memory, spatial skill, verbal logic, speed — not a loose list of talents in general.

Eight domains MOSAIC-48 measures

MOSAIC-48 organizes its items into eight domains, each mapping onto a broad or narrow ability in the CHC tradition. Where the mapping is well established, the standard CHC code is noted in parentheses; a couple of the domains describe a control process or a narrow ability that sits under a broader one rather than a broad ability in its own right.

• Fluid Reasoning (Gf) — discovering and applying rules in problems you have never seen; the broad ability that abstract matrices and sequences are built to tap.
• Visual-Spatial (Gv) — rotating, folding, and re-viewing forms in the mind, from mental rotation to reading a shape from a new angle.
• Working Memory (Gsm) — holding a handful of items in mind while you rearrange or transform them; the mental workspace that reasoning runs on.
• Quantitative Reasoning — seeing structure in numbers, ratios, and uncertainty; CHC treats numerical reasoning as a narrow ability leaning on fluid reasoning rather than on memorized arithmetic.
• Learning Efficiency (Glr) — picking up a rule you were just taught and retrieving it accurately a little later; the acquisition-and-retrieval side of ability.
• Verbal Reasoning — tracking logical and conceptual structure carried in words, which draws on reasoning applied to language rather than on vocabulary trivia.
• Attention Control — inhibiting an obvious-but-wrong pull and switching rules on cue; an executive control process that supports the other abilities rather than a store of knowledge.
• Processing Speed (Gs) — comparing simple information quickly and accurately under mild time pressure.

Why the distinctions matter

Because these abilities are only partly correlated, a strength in one does not guarantee a strength in another. Someone can reason fluidly yet find timed speed tasks draining, or hold a rich vocabulary yet work at an average memory span. Seeing the domains separately turns "how smart are you?" into the more useful "smart at what?"